The Standard Normal (z) Table
Cumulative left-tail area under the standard normal curve: the probability that an observation falls at or below a given z-value.
Statistical Tables
The four statistical tables from Making Sense of Social Data, readable in a browser. They are not printed in the book: this is where they live. Twenty pages of print became a page you can open in a lab, on a phone, or beside the exercise you are working on.
The t Distribution Table
Two-tailed critical values of the Student t distribution, by degrees of freedom and significance level. Used for confidence intervals and tests of means.
The Chi-Square Table
Right-tail critical values of the chi-square distribution. The chi-square test is always right-tailed, so only that tail is tabulated.
The F Distribution Table
Right-tail critical values of the F distribution at α = .05 and .01, for analysis of variance and for comparing two variances.
Reading a wide table on a phone
These tables are wide — the z table is ten columns and the F table thirteen. On a narrow screen each one scrolls sideways inside its own box rather than stretching the whole page, and the left-hand column of row labels stays in place while you scroll, so the row you are reading across is never unlabelled. If you are looking up a lot of values, the PDF and a larger screen will be less work.
Why these left the book
They cost twenty pages, and they are the part of a statistics textbook a reader consults rather than reads. Measured: the book compiles to 725 pages with the four tables and 705 without.
Being online is also the better place for them. A table you need during a lab, or beside an exercise on a laptop, is easier to reach in a browser tab than by flipping to the back of a 700-page book — and unlike the printed version, this one cannot be left at home.
There is one source and there are two outputs. The PDF and these
pages both come from the same four files the book used to print, and
every cell on this site is computed from R's qnorm(),
qt(), qchisq() and qf() rather
than copied. The generator re-checks all 1,924 published values
against those functions each time it runs, and refuses to publish if
one disagrees.
Found an error?
All 1,924 values have been checked against R and none of them disagreed. If you find one that does, tell me — that would be a real bug and I would want to know today.